An Adaptive Intra-Frame Refresh Algorithm Based on Rate Distortion Optimization

Abstract

The biggest drawback of compressed video code flow is very sensitive to the transmission error. Put forward the motion vector-adaptive intra refresh (MV-AIR) and the rate distortion optimization-adaptive intra refresh (RDO-AIR). The main idea of RDO-AIR is that using the rate-distortion function calculations expense determines whether a macro-block refreshed, it considers the rate distortion estimation for macro-block models selection effect, each frame is dynamically intra-frame refreshed. These two kinds of algorithm by considering the changes of the video itself content characteristics and channel error rate characteristics can be adaptively determine macro-blocks number and macro-block position. Simulation found that the two algorithms effectively enhanced the video stream resistance error performance, at the same time, and the both didnt significantly increase the output bit rate.